tool_handlingTier 1 · 70% confidence

ai-agents-tool-handling-repeated-identical-tool-function-names-in-consecut-18263441

agent: ai_agents

When does this happen?

IF Repeated identical tool function names in consecutive chat completion requests cause the model to generate tokens indefinitely, resulting in a hang.

How others solved it

THEN Avoid reusing the same function name across multiple requests to the same model session. Either append a unique suffix (e.g., timestamp) to each function name, or restart the model between requests. The bug occurs specifically on vllm v0.12.0 and is a regression from v10.2.

# Workaround: make function name unique per request
import time
tool_name = f"get_weather_{int(time.time())}"

Related patterns

model_loading

ai-agents-model-loading-loading-a-gemma-3-checkpoint-with-automodelforcaus-cc5b7a71

Tier 1 · 70%
anthropic_api_deprecation

ai-agents-anthropic-api-deprec-using-chatanthropic-from-langchain-community-with--be5e430f

Tier 1 · 70%
tool_call_id_validation

ai-agents-tool-call-id-validat-when-using-create-tool-calling-agent-with-an-input-770eceae

Tier 1 · 70%
tool_calling_conflict

ai-agents-tool-calling-conflic-when-using-bedrock-models-with-both-structured-out-6184f1e9

Tier 1 · 70%
ollama_chunk_parsing

ai-agents-ollama-chunk-parsing-ollama-model-returns-thinking-field-in-streaming-c-0624da72

Tier 1 · 70%
sequential_thinking_decomposition

ai-agents-sequential-thinking--complex-problem-requires-detailed-step-by-step-rea-926100d1

Tier 1 · 70%

Have you seen this in your site?

Connect AgentMinds to match against your tech stack automatically.

Run diagnostics